Re: Feature questions

[Bernino Lind]

Dear Sirs,

Thank you very much for your kind reply.

The A+ link is unfortunately not really an option since I need GUI tools
like the ones presented in GNUe. Corporate policies...

In general I would like to give you guys my opinion or experience. I work
with business management and consulting, so what I say and hear is very
much a reflection of the thoughts, trends and fashion in top-management of
big corporations.

Let me give a historic overview which is probably the most consistent way
of telling the story of IT and business. The years would ofcourse be
offset +-10 years or so depending on type of business, size etc. the
general point being the evolution...

30 years ago: introduction of PDP's etc. to do automated processes such as
recurring invoicing etc. No storage of transaction in digital form.

20 years ago: first databases occur and the automated processes
(transactions) gets stored in digital form. Usage of storage is just like
previously documentation.

15 years ago: data from databases starts to get used in reports. We sold
this and that in month x and y. More processes gets digitized => more data
to present/report on.

10 years ago: Marketing people, engineers etc. start to use the historical
data - "we saw an increase of sales after doing this and that".

5 years ago: Business Intelligence - balanced scorecards OLAP (online
analytical processing), the historical data from various operational
databases gets "copied" into the OLAP for online on the fly reporting.
Sales yesterday can be seen now, correlated with geographic data, gender,
...., all the data in our *data warehouse*.

Balanced scorecards is a way to present the business values and processes
in a homegenic form which can be used to measure the perfomance of the
business. Really one tries to measure things such as "customer
satisfaction" and put up threshold values...how this is possible is a very
tedious and long story.

Today: Forecasting - or some call it analytical intelligence. The complex
array of historical data that has been B.I.'ed to extract "knowledge" of
the data is now used to predict tomorrow correlated with parameters that
influence what ever we want to predict.

Example: inventory of a bakery - how much butter, flour etc. etc. should
we buy for the production tomorrow?

----

Tools for this: all the relevant ERP must be there (generating data),
reporting tools, data warehouse tools (ETL process and catching data from
many sources), statistical and analytical tools.

Oracle and SAS I would say are market leaders in this area: tip, SAS has a
"student tool package" to get inspiration...

Best regards, Bernino Lind





> Am Don, 2003-04-03 um 22.43 schrieb Derek Neighbors:
> <snip>
>> > 2. Which sources can Integrator read?
>> Currently none as Integrator has been officially started.  That said
>> you could use our reports engine to do some interesting things.
> There is a working implementation in the integrator directory in CVS.
> But there is still missing an official Plan or Featurelist for
> integrator. (Derek, could you please :) )
>
>> > 2.1 Where does it store the data it reads?
>> > 2.2 Does it transform it to XML?
>> The answer will be yes it will store into XML.  Whether you write it
>> as XML or it only deals with XML in memory likely will depend on your
>> transformations needs.
>>
>> > 2.3 Can it read flat files like excel documents and access
>> databases?
>> That is one of the requirements.
> The actual,inofficial implementation uses the db abstraction layer
> provided by GNUe Common, which provides access to the most common
> databases and f.e. DBase III (dbf) files. (CSV file access is being
> worked on.) The official Integrator will probably use GNUe Common too.
> <snap>
>
> Jan
>
> --
> Jan Ischebeck <address@hidden>
>
>
> _______________________________________________
> Gnue mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gnue